An AI assistant that helps you understand YouTube videos. Paste a link and ask questions, get summaries, or generate chapter timestamps — all backed by the actual transcript with clickable citations. Built with Mastra.
This template shows how to build a conversational agent with persistent memory that uses custom tools to fetch and analyze external content. It demonstrates tool composition — getting metadata first for context, then fetching the transcript for detailed analysis.
- Summarize video content with timestamped sections
- Answer questions about what was said in the video
- Generate chapter timestamps for easy navigation
- Find specific moments where topics are discussed
- Clickable timestamp citations that link directly to the video
- Conversational memory to ask follow-up questions
- OpenAI API key — used by default, but you can swap in any model
- Clone the template
- Run
npx create-mastra@latest --template chat-with-youtubeto scaffold the project locally.
- Run
- Add your API keys
- Copy
.env.exampleto.envand fill in your OpenAI key.
- Copy
- Start the dev server
- Run
npm run devand open localhost:4111 to try it out.
- Run
Open Studio and start chatting with the chat-with-youtube agent. Paste any YouTube URL and ask for a summary, specific questions about the content, or chapter timestamps. The agent fetches the transcript and responds with citations you can click to jump to that moment in the video.
Swap in a different model, add your own tools for video analysis, or wire the agent into your app using the Mastra Client SDK. The agent and tools are all in src/ — edit them directly to fit your use case.
Mastra templates are ready-to-use projects that show off what you can build — clone one, poke around, and make it yours. They live in the Mastra monorepo and are automatically synced to standalone repositories for easier cloning.
Want to contribute? See CONTRIBUTING.md.